Dear SPM experts,
I recently posted a question but maybe it was not clear enough:
I did two 2nd-level analyses: in both I have the factor stress with 2 levels (stress vs control) but the 2nd factor differs between the analyses; in my first analysis, the 2nd factor is ADRA2B genotype with 2 levels (del vs wt carriers) with equally sized groups; in my second analysis, the 2nd factor is MR-Haplotype with 2 levels (carriers vs non-carriers) with unequal group sizes (control carrier:45, stress carrier:43, control non-carrier:15, stress non-carriers:13).
In my results I am looking at the main effect of stress, so although group sizes differ between the analyses, the analysis for the main effect of stress of course comprise the same number of participants. However, whereas in the 1st analysis I find sig. greater functional connectivity between the amygdala and the hippocampus in the control condition, this effect is no longer sig. in my 2nd analysis!
Does anyone know how spm calculates main effects so that the effects differ?
Thanks a lot in advance!
Best wishes,
Lisa W.
|